ODE: Newton's Law of Cooling

Newton's Law of Cooling provides a powerful yet simple mathematical model for predicting how the temperature of an object changes when it is immersed in a surrounding medium. As a classic application of first-order ordinary differential equations (ODEs), it bridges abstract calculus and tangible engineering and forensic problems, from designing thermal systems to estimating the time of death in criminal investigations. Mastering this model teaches you how to translate a physical observation into a solvable differential equation and extract meaningful predictions from its solution.

Derivation of the Law

The law is based on a core empirical observation: the rate at which an object's temperature changes is proportional to the difference between its own temperature and the ambient temperature of its surroundings. This is a statement about the derivative of temperature with respect to time.

Let $T(t)$ represent the object's temperature at time $t$, and let $T_s$ represent the constant temperature of the surrounding environment. The observation translates directly into the differential equation:

$$\frac{dT}{dt}=k(T-T_s)$$

Here, $k$ is the constant of proportionality, known as the cooling constant. Its value depends on the physical properties of the object and its environment (e.g., surface area, material, airflow). Crucially, if the object is cooling down (i.e., $T>T_s$), the temperature derivative $\frac{dT}{dt}$ should be negative. This forces $k$ to be a negative number for a cooling process. Conversely, if the object is heating up in a warmer environment ($T<T_s$), then $k$ would be positive. We often write the law with the negative sign explicit: $\frac{dT}{dt}=-k(T-T_s)$, where $k>0$ for cooling.

The General Solution

The equation $\frac{dT}{dt}=k(T-T_s)$ is a first-order linear ODE. It is also separable, making it straightforward to solve.

1. Separate Variables: Rearrange the equation to group $T$ and $t$ terms:

$$\frac{dT}{T-T_s}=kdt$$

1. Integrate Both Sides:

$$\int \frac{dT}{T-T_s}=\int kdt$$ This yields $\ln |T-T_s|=kt+C$, where $C$ is the constant of integration.

1. Solve for T(t): Exponentiate both sides to remove the logarithm:

$$|T-T_s|=e^{kt+C}=e^C{e}^{kt}$$ Let $A=\pm e^C$ (absorbing the absolute value). The general solution is: $$T(t)-T_s=A{e}^{kt}\text{or}T(t)=T_s+A{e}^{kt}$$

1. Apply Initial Condition: Suppose we know the initial temperature $T(0)=T_0$. Substituting $t=0$ gives $T_0=T_s+A{e}^0=T_s+A$, so $A=T_0-T_s$. The final, particular solution is:

$$T(t)=T_s+(T_0-T_s)e^{kt}$$

This elegant formula shows the temperature difference between the object and its surroundings decays exponentially over time. For a cooling process ($k<0$), the exponential term $e^{kt}$ decays to zero, and $T(t)$ asymptotically approaches $T_s$.

Example: A cup of coffee at $95^{\circ }$C is placed in a $20^{\circ }$C room. If the cooling constant is $k=-0.1{\text{min}}^{-1}$, its temperature after 10 minutes is: $$T(10)=20+(95-20)e^{-0.1\cdot 10}=20+75e^{-1}\approx 20+27.6=47.6^{\circ }\text{C}.$$

Determining the Cooling Constant k

Often, $k$ is not given but must be determined from experimental data. This involves taking temperature measurements at known times and solving for $k$.

Starting from the solution $T(t)=T_s+(T_0-T_s)e^{kt}$, you can isolate $k$: $$\frac{T(t)-T_s}{T_0-T_s}=e^{kt}$$ Taking the natural logarithm of both sides: $$\ln \left(\frac{T(t)-T_s}{T_0-T_s}\right)=kt$$ Therefore, $$k=\frac{1}{t}\ln \left(\frac{T(t)-T_s}{T_0-T_s}\right)$$

You only need one data point $(t,T(t))$ along with $T_0$ and $T_s$ to calculate $k$. For better accuracy, multiple data points can be used with linear regression: plotting $\ln |T(t)-T_s|$ versus $t$ should yield a straight line with slope $k$.

Advanced Applications and Limitations

Forensic Application: Time of Death Estimation

This is a classic "inverse problem." A medical examiner arrives at a crime scene to find a body. The ambient temperature $T_s$ is known (e.g., room temperature). The body's current temperature $T$ is measured. Assuming a normal living body temperature of $T_0\approx 37^{\circ }$C (98.6${}^{\circ }$F), and using an established average cooling constant $k$ for a human body in those conditions, the time since death $\Delta t$ can be estimated.

Rearranging the solution formula: $$T=T_s+(37-T_s)e^{k\Delta t}$$ Solving for $\Delta t$: $$\Delta t=\frac{1}{k}\ln \left(\frac{T-T_s}{37-T_s}\right)$$

Crucial Note: This estimate depends heavily on the accurate value of $k$, which varies with body size, clothing, air humidity, and whether the body was submerged. It is most reliable for the first few hours after death and is always used as one piece of evidence among many.

Limitations for Large Temperature Differences

Newton's Law of Cooling is an excellent model when the temperature difference $\Delta T=|T-T_s|$ is not too large and heat transfer occurs primarily via convection. Its core limitation is that it assumes the cooling constant $k$ is truly constant. For very large temperature differences:

• The dominant mode of heat transfer may shift from convection to radiation, which depends on $T^4-T_s^4$, not simply $(T-T_s)$.
• Material properties (like thermal conductivity or specific heat) may change with temperature.
• Significant evaporation or phase changes may occur (e.g., a boiling pot of water).

In such cases, the simple linear ODE model becomes less accurate, and a more complex, often nonlinear, model is required.

Common Pitfalls

1. Sign Error on k: The most frequent mistake is forgetting that $k$ is negative for a cooling object. If you use a positive $k$ in $T(t)=T_s+(T_0-T_s)e^{kt}$, your model will show the object heating up exponentially instead of cooling down. Correction: Remember the physical meaning. If $T_0>T_s$ (cooling), then $\frac{dT}{dt}<0$, so $k$ must be negative. You can write the law as $\frac{dT}{dt}=-K(T-T_s)$ with $K>0$ to avoid this.

1. Misidentifying the Surrounding Temperature $T_s$: $T_s$ must be the constant ambient temperature. Do not confuse it with an initial temperature or a final temperature. If the surroundings are changing temperature (e.g., a cooling object placed in a warming room), the basic model does not apply without modification. Correction: Carefully read the problem statement. $T_s$ is typically described as "room temperature," "ambient temperature," or "the temperature of the bath."

1. Incorrect Application of the Time-of-Death Formula: Using the formula $\Delta t=\frac{1}{k}\ln \left(\frac{T-T_s}{37-T_s}\right)$ without critical thought. If the measured body temperature $T$ is below the ambient temperature $T_s$, the argument of the logarithm becomes negative, which is impossible—indicating the model has broken down or the ambient temperature has changed. Correction: Understand that this model is valid only while the body is cooling toward $T_s$. It cannot predict times for a body that has reached ambient temperature.

Summary

• Newton's Law of Cooling states the rate of temperature change is proportional to the difference between the object's temperature $T$ and the ambient temperature $T_s$, leading to the ODE: $\frac{dT}{dt}=k(T-T_s)$.
• The solution is an exponential decay of the temperature difference: $T(t)=T_s+(T_0-T_s)e^{kt}$, where $k<0$ for cooling.
• The cooling constant $k$ can be determined experimentally using the relation $k=\frac{1}{t}\ln \left(\frac{T(t)-T_s}{T_0-T_s}\right)$.
• A key application is forensic time-of-death estimation, though it requires careful consideration of the value of $k$ and the model's assumptions.
• The model has limitations, primarily its assumption of a constant $k$. It becomes less accurate for very large temperature differences where radiation becomes significant or material properties change.